roux/scripts/import-posts.py

#!/usr/bin/env python3
"""
Import posts from ~/projects/ginger/posts.csv into Hugo content.
Only imports posts whose generated_image exists in the selected images folder.
Run from the site root: python3 scripts/import-posts.py
"""
import csv
import os
import shutil
from pathlib import Path

GINGER   = Path.home() / "projects" / "ginger"
CSV_PATH = GINGER / "posts.csv"
IMG_DIR  = GINGER / "images" / "final" / "selected"
SITE     = Path(__file__).parent.parent
CONTENT  = SITE / "content" / "posts"

CONTENT.mkdir(parents=True, exist_ok=True)

# Collect available images
available = {f.name for f in IMG_DIR.iterdir() if f.is_file()}

matched = []
with open(CSV_PATH, newline="", encoding="utf-8") as f:
    reader = csv.DictReader(f)
    for row in reader:
        gen = row.get("generated_image", "").strip()
        if gen and gen in available:
            matched.append(row)

print(f"Matched {len(matched)} posts out of {len(available)} available images")

# Sort deterministically by generated_image filename for stable plate numbers
matched.sort(key=lambda r: r["generated_image"])

for idx, row in enumerate(matched, start=1):
    plate    = f"{idx:03d}"
    gen_img  = row["generated_image"].strip()
    slug     = gen_img.replace(".png", "")
    title    = row["title"].strip()
    desc     = row["description"].strip()
    if desc:
        desc = desc[0].upper() + desc[1:]
        if not desc.endswith('.'):
            desc += '.'
    desc = desc.replace('"', '\\"')
    raw_cat  = row.get("category", "").strip()
    raw_tags = row.get("tags", "").strip()

    cats = [raw_cat.split(",")[0].strip()] if raw_cat.strip() else ["Uncategorised"]
    tags = [t.strip() for t in raw_tags.split(",") if t.strip()]

    cats_yaml = "\n".join(f'  - "{c}"' for c in cats)
    tags_yaml = "\n".join(f'  - "{t}"' for t in tags) if tags else '  []'

    bundle  = CONTENT / slug
    bundle.mkdir(exist_ok=True)

    src_img = IMG_DIR / gen_img
    dst_img = bundle / gen_img
    if not dst_img.exists():
        shutil.copy2(src_img, dst_img)

    md = f"""---
title: "{title.replace('"', '\\"')}"
description: "{desc}"
plate: "{plate}"
slug: "{slug}"
issues:
  - "01"
image: "{gen_img}"
weight: {idx}
categories:
{cats_yaml}
tags:
{tags_yaml}
---
"""
    (bundle / "index.md").write_text(md, encoding="utf-8")

print(f"Created {len(matched)} page bundles in {CONTENT}")